The goals of this project are to detect and accurately describe perimenopausal mood disorders, explore their pathophysiology and response to pharmacological and environmental manipulation, and to document the relationship between reproductive endocrine change and disorders of mood as a way of further investigating the neurobiology of psychiatric illness. Findings to date include: 1) no diagnosis-related differences in the basal plasma levels of FSH, LH, estradiol, estrone, and total or free testosterone were identified in 21 women with depression during the perimenopause and 21 asymptomatic age and reproductive status matched controls; 2) significantly decreased morning basal plasma levels of the adrenal androgen DHEA in depressed perimenopausal women compared to controls; 3) significant improvement in multiple symptoms and standardized rating scale scores of depression and in measures of verbal memory following estrogen replacement in women with perimenopausal depression independent of estrogen's salutary effects on vasomotor symptoms; 4) the recurrence of depressive symptoms after the administration of the serotonin agent metergoline but not placebo in six depressed perimenopausal women who responded to estrogen replacement under placebo-controlled conditions. Additionally, in an experimental paradigm involving the induction of hypogonadism in men and women with GnRH agonists we have observed the hypogonadal state to be associated with the following: 1) decreased measures of sexual functioning in both men and women with a restoration of normal sexual function during gonadal steroid replacement in men with testosterone but not in women with either estradiol or progesterone; 2) no significant effects on neuropsychological test scores in young women, but improved visual spatial performance in men; 3) decreased exercise-associated pleasure in men; 4) elimination of cognition activated regional cerebral blood flow (O15 PET scans) in the dorsolateral prefrontal cortex in women but not men; 5) delayed onset of the melatonin secretory curve and increased amplitude of melatonin secretion in women but not men; 6) increased m-CPP stimulated growth hormone secretion in women compared to men; 7) increased measures of platelet Gi and PKC alpha isoenzyme activity in both men and women; and 8) no gonadal steroid-related differences in cerebral spinal fluid measures of monoamine metabolites in men. Finally, we have observed that progesterone but not estrogen replacement in women is associated with an increased m-CPP-stimulated secretion of prolactin (area under the curve) and a decreased maximum stimulated core temperature relative to baseline, compared to the hypogonadal state. No gonadal steroid-related changes in m-CPP-stimulated hormonal or core temperature measures were observed in men.